The goal of the proposed study is to investigate within-family differences in parent-child relations in later life, and the consequences of these differences for parents and their children. In 2001, we initiated the first large- scale study of within-family differences in parent-adult child relations: The Within-Family Differences Study (WFDS-I). Data were collected from 566 mothers ages 65-75, 773 of their adult children, and 130 of their husbands. Information about parents'relationships with each of their adult children was collected from mothers and fathers;in 300 families, data were collected from members of both generations. The WFDS-I revealed that most parents differentiate among their adult children, and that these patterns can be predicted by social structural and value similarity, intergenerational exchange, and developmental factors. The WFDS-I Was ideally suited to investigate patterns of within-family differentiation in later life;however its cross- sectional design did not allow examination of changes across time in differentiation or the consequences of such differentiation. In this application, we propose conducting a second wave of data collection, in which parents and adult children will be interviewed approximately five years after they participated in the original study. This design will allow us to address three specific aims: 1) Examining the ways in which patterns of within-family differences in parent-child relations change across time in response to events in parents'and children's lives. 2) Studying the transition to caregiving among adult children, including factors that predict why some children assume greater caregiving responsibilities, and how outcomes for both parents and children are affected by which children become caregivers. 3) Understanding of the effects of children's problems on parents'well-being and how this relationship is moderated by within-family differences in both the occurrence and characteristics of such problems. An oversample of Black families allows for consideration of race throughout the analyses. The proposed project has clear relevance to public health issues. The study will provide new data regarding the role of family relationships as predictors of physical and mental health in later life. Further, it will shed new light on the nature and dynamics of the provision of care to older persons by adult children;this information is likely to be useful in developing and refining interventions to assist both caregivers and care recipients.